摘要
提出了一种改进的粒子群算法,在初始化种群时采用相对基学习原理,以获得较优的初始候选解;在后期迭代过程中引入扩张模型,使粒子不易陷入局部极小值点,并将其用于多阈值图像分割。由最大熵阈值法得到所要优化的目标函数,用改进的粒子群算法对其进行优化,使其能够准确并迅速的得到分割的最佳阈值组合,并用该阈值组合对图像进行分割。将此分割结果与遗传算法的多阈值分割结果相比较可以看出,该算法可更为准确快速的实现图像分割。
To determine the optimal thresholds in image segmentation,an improved particle swarm optimization(pso) is put forward. In this method, it adopted Opposition-Based Learning in initialization to get a better solution and adopted expansion model in later iteration to avoid getting into local minumum.The optimization object function using maximum entropy(ME)method can be gotten. By the optimization of the object function, the optimal thresholds can be gotten well and quickly, and the image by use of the thresholds can be segmented. Compared with the Genetic Algorithm (GA), the experimental results show that the improved PSO realizes the image segmentation well and quickly.
出处
《微型电脑应用》
2011年第5期59-61,70,共4页
Microcomputer Applications
关键词
粒子群优化算法
相对基学习
扩张模型
多阈值
图像分割
Particle Swarm Optimization
Opposition-Based Learning
Expansion Model
Multilevel Thresholding
Image Segmentation
